AI Dad
All records

Genspark · web_search

Grade 2

Works in both Korean and English. All 25 returned results, and the same keyword gives the same answer on a repeat run.

Pull the first page of search results for 25 keywords

  1. Runs
  2. Records
  3. Detects
  4. Precise
  5. Proven

25 / 25keywords answered

What failed It does not return the ten results asked for — between six and ten depending on the keyword (stable within a keyword). It never says why the rest are missing.

How we measured How we measured precision: three keywords (two English, one Korean), three runs each, back to back on the same day, comparing the ranked list. All nine comparisons matched exactly, order included. Every run returned nine results.

Not measured We only repeated within a single day. How far results drift across days is still unknown. We also did not open the ranked pages to confirm they are what they claim (the Proven rung).

2026-09-09 · gsk CLI, as of 2026-09-09

Rung by rung

  1. Runs

    Passed

    It runs. Install it, run it, something comes out.

  2. Records

    Passed

    It leaves a trace you can check later.

  3. Detects

    Not measured

    It says when something is wrong instead of passing silently.

  4. Precise

    Passed

    Same input, same answer. It does not drift.

  5. Proven

    Not measured

    Verified on the real thing — output seen with our own eyes.

How to read this

This result came from the task named above, on the date named above, against the version named above. A different task gives a different result. "Not measured" means we have not checked it — not that the tool cannot do it.